Progression of Acromegalic Arthropathy in Long-Term Controlled Acromegaly Patients

9 Years of Longitudinal Follow-Up

Iris C.M. Pelsma; Nienke R. Biermasz; Wouter R. van Furth; Alberto M. Pereira; Herman M. Kroon; Margreet Kloppenburg; Kim M.J.A. Claessen


J Clin Endocrinol Metab. 2021;106(1):188-200. 

In This Article


Patient Characteristics

Thirty-one patients with acromegaly in long-term remission were included. Clinical baseline characteristics of the population are shown in Table 1. Median age at baseline was 60 years (IQR, 53–66 years) and 49% were women. At baseline, 18 patients (58%) were in biochemical remission after transsphenoidal surgery, with or without additional radiotherapy, and 13 patients (42%) received adjuvant treatment with SMS analogues, initiated after surgery or radiotherapy in all patients. None of the included patients were treated with PegV. During the study, SMS analogues were discontinued in 2 patients, whereas in 1 patient SMS analogues were started for recurrent disease activity after previous surgical remission.

Structural radiographic joint abnormalities were very common, with radiographic OA in 1 or more joints in all patients at baseline, as shown in Table 1. At baseline, 2 patients had a unilateral hip prosthesis, and 1 patient had bilateral knee prostheses. No patients were diagnosed with rheumatoid arthritis.

Course of Joint Symptoms in Time

Self-reported Pain and Disability. Lower Limb: Median total and subscale WOMAC scores did not change over time, as depicted in Table 2. As depicted in the cumulative probability plots in Figure 1, individual scores varied greatly. Clinically relevant deterioration was found only in a small subset of patients (15%; 23%; 13% for the pain, stiffness, and function subscales), whereas the remaining patients reported unchanged (70%; 46%; and 74%, respectively) or improved scores for the respective subscales (15%; 31%; and 13%, respectively) (see Table 2).

Figure 1.

Cumulative probability plots of the change in self-reported pain, stiffness, and functional disability of the lower limb. Individual change scores for the A, pain, B, stiffness, and C, function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) questionnaire during 9.1 years of follow-up are depicted in black circles for 27, 26, and 23 patients, respectively. The horizontal dashed lines represent the cutoff for deterioration (upper line) and improvement based (lower line) on the minimum perceptible clinical improvement (MPCI) for the respective subscales (9.7 mm for pain, 10.0 mm for stiffness, and 9.3 mm for function) (45). CPP, cumulative probability plot.

Hands: Median total AUSCAN scores significantly deteriorated during follow-up, despite large interindividual variety (see Table 2), which was driven by a deterioration in hand function. In Figure 2, cumulative probability plots with absolute AUSCAN change scores illustrate a left-skewed distribution of change in pain and function scores over time. With respect to pain, clinically relevant deterioration was observed in 32% of patients, whereas the remaining patients were classified as unchanged (52%) or even improved (6%). For the hand function subscale, 39% of the patients deteriorated over time, whereas 43% remained unchanged and 18% improved (Table 2).

Figure 2.

Cumulative probability plots of the change in self-reported pain and functional limitations of the hand. Individual change scores for the A, pain and B, function subscales of the Australian/Canadian Osteoarthritis Index (AUSCAN) questionnaire during 9.1 years of follow-up are depicted in black circles for 28 patients. The horizontal dashed lines represent the cutoff for clinical progression (upper line) and improvement (lower line), based on the minimum clinically important improvement (MCII) (1.49 and 1.25 for the pain and function subscales, respectively) (44). CPP, cumulative probability plot.

Performance Tests. Lumbar flexion index did not change (14.0 ± 1.1 to 13.9 ± 0.9, P = .73), whereas cylinder grip strength deteriorated over time in both hands (right: 38.4 ± 15.2 to 25.0 ± 13.7, P = .02; left: 36.7 ± 14.2 to 33.0 ± 13.6, P = .006). Notably, of all the patients, 23 patients were right-handed, 1 was ambidextrous and 3 were left-handed; for 4 patients, dextrality was unknown.

Radiographic Osteoarthritis Course in Time

Hips. Median KL scores for the hips, increased significantly over time (0 IQR, 0–4 vs 3 IQR, 0–6; P < .001; Figure 3A). Fifteen patients (48%) showed radiographic progression during total follow-up, resulting in a total hip prosthesis in three patients (1 bilateral joint replacement, 2 unilateral prosthesis). Progression occurred in 7 patients (23%) during the first 2.6-year follow-up period, and in an additional 8 patients (26%) during the second 6.7-year follow-up period.

Figure 3.

Radiographic progression for the different joint sites during 9.1 years of follow-up. A and B, Kellgren and Lawrence score (KL) scores (median [interquartile range]; squares) for the A, hip, and B, knee joints at the 3 study visits (baseline, after 2.6 and 9.1 years of follow-up, respectively) are depicted. Differences in KL scores between the 3 study visits are calculated using the Friedman test. C-F, KL scores (mean ± SD, circles) for C, hand, D, axial, E, facet osteoarthritis, and F, disc degeneration, are shown for the aforementioned study visits. Axial OA is defined as a combination of facet OA and disc degeneration. Differences in KL scores between the study visits are calculated using repeated-measures analysis of variance.

Knees. Median KL scores for the knees increased significantly over time (1 IQR 0–3 vs 3 IQR, 0–5; P < .001; Figure 3B). Radiographic progression was observed in 9 patients (29%) during the entire study period, of whom 1 patient received a unilateral joint replacement. All patients showed progression during the second follow-up period.

Hands. Mean KL scores for the hands increased significantly over time (20 ± 16 vs 28 ± 18, P < .001; Figure 3C), with hand OA progression being observed in 26 patients (84%). In 8 patients (26%), progression occurred during the first follow-up period, whereas in 25 patients (81%) progression was observed during the second follow-up period.

Spine. Mean KL scores for either facet OA (17 ± 5 vs 20 ± 5, P < .001), disc degeneration (10 ± 6 vs 13 ± 7, P < .001), or combined axial OA (26 ± 9 vs 33 ± 10, P < .001) increased significantly during follow-up (Figure 3D-3F). Progression of facet OA was observed in 25 patients (81%) during the entire study period (N = 18; 58% in the first study period; N = 23; 74% in the second study period) and disc OA progression was observed in 19 (61%), 10 (32%), and 13 patients (42%) during the total, first, and second follow-up periods, respectively. Moreover, progression of combined axial OA was observed in 29 patients (94%), with 24 patients (77%) showing progression both in the first follow-up and the second follow-up periods.

Risk Factors for Radiographic Progression

Progression of hip OA was significantly associated with higher baseline hip KL scores (odds ratio [OR] 1.88; 95% CI 1.09–3.16), P = .02) adjusted for age and sex (Table 3), but not with active disease duration, treatment modality, pretreatment IGF-1 levels, nor after correction for baseline KL scores. Progression of knee, hand and axial OA was not associated with the aforementioned factors.

Relationship Between Clinical and Radiographic Osteoarthritis Progression

Mean change in self-reported AUSCAN stiffness scores was positively associated with the change in KL scores for the hands (ρ = 0.456, P = .02), as summarized in Table 4. Moreover, mean change in WOMAC stiffness scores showed an insignificant trend toward a positive association with the change in KL scores for the knee (ρ = 0.362, P = .07; see Table 4). Other AUSCAN and WOMAC subscales were not associated with radiographic progression of hand and lower limb OA, respectively.